NAEP Technical DocumentationStratification Variables

Stratification by Urbanization Classification

Stratification by Minority Classification

Stratification by Achievement Data and Median Income

Missing Stratification Variables

The stratification of public schools for the 2003 state assessment involved four dimensions. Public schools were stratified hierarchically by the following variables:

classification as a charter school (fourth grade only) or not;
urban classification, combined with whether or not school is in a Trial Urban District Assessment (TUDA) district;
minority classification; and
achievement score or median income of the ZIP Code area.

For the fourth-grade public schools, the initial stratification variable denotes whether or not the school was classified as a charter school. Fourth-grade charter schools in some jurisdictions were sampled at higher rates than other schools.

The urbanization variable classifies schools by up to eight types of location (e.g., a center city, an urban fringe of a city, a town, or a rural area), and also classifies as to the size of the city or town, and whether a rural area is in a Metropolitan Statistical Area. This variable was combined with an indicator of whether the school was located in a TUDA district.

Minority status variables classify schools on percentages of minorities found in the school which are predominant in particular jurisdictions. Minority classification is based on the Common Core of Data (CCD) race/ethnicity variable and can be defined to be any group that represents a minority in a given jurisdiction. The minority cells are nested within each cell defined by the urbanization classification (within each jurisdiction). Jurisdiction-urbanization cells with no predominant minorities (less than 7 percent for every minority) have no minority classifications. Jurisdiction-urbanization cells with two predominant minorities (greater than 15 percent for at least two minorities) have four minority cells based on percentages of the two predominant minorities. Jurisdiction-urbanization cells with one minority greater than 7 percent, but less than 15 percent, and one at most greater than 15 percent have three minority cells based on the sum of the percentages of the two most prevalent minorities.

The last stratification variable is achievement score or median household income. Achievement data were used if they were available for a particular jurisdiction and grade. If achievement data were not available, median household income of the ZIP Code area where the school was located was used.

The implicit stratification in this four-fold hierarchical procedure was achieved via a "serpentine sort." This sort was accomplished by alternating between ascending and descending sort order on each variable successively through the sort hierarchy. Schools were sorted first by charter school status (for fourth grade schools). Within the two categories of charter school status, schools were arranged, in serpentine order, by urbanization/TUDA classification, with ascending order in the first charter school cell and descending order in the second charter school cell. Within each urbanization/TUDA classification, minority classification cells were sorted by ascending order for one urbanization/TUDA classification, followed by descending order for the next urbanization/TUDA classification, and so on. Finally, within each minority classification cell, the schools were arranged, in serpentine order, by achievement data or median household income, with ascending order for achievement data/median household income used in every other cell and descending order for achievement data/median household income used in the remaining cells, giving an ascending-descending-ascending-descending pattern. Because there was no oversampling of eighth-grade charter schools, the hierarchy for eighth grade did not include charter school status.

Last updated 02 October 2008 (KL)

Printer-friendly Version